Maps of Meaning and Peter Jackson

Right out of the box let me note that I did not read even the majority of Peter Jackson's often intricate arguments and review of other people's research in his book Maps of Meaning: An Introduction to Cultural Geography, New York (Routledge) 1989. But here are a couple of thoughts that stuck with me when leafing through the book. 

There were successful combinations of spatial and sociological arguments, such as David Ley's analysis in The Black Inner City as Frontier Outpost (= Monograph Series no. 7), Washington, DC (Association of American Geographers), 1974, of inter-gang violence in Philadelphia, that marked out the territories of the gangs, overlaid them on a map, and then showed with lines of various thickness where the hotspots of violence occurred (chapter Culture and Ideology, section Territorial Struggles, pp.68-70; map on p.71); or the collaboration between Ley and R. Cybriwski, Urban graffiti as territorial markers, in: Annals of the Association of American Geographers, vol 64 (1974), pp.491-505, which argued that graffitis are used as markers in contested border areas (pp.70; map on p.73), previously identified by the gang-violence analysis. 

Thus, arguments of spatial correlation of events are possible that would be hard to spot in other representations. The problem is that the objects that are identified in the environment are configurations of physicality, but the mere mapping of the physical is insufficient. As Jackson puts it in his summary,

Too often in the past, cultural geography has degenerated into the an inventory of 'culture traits', mapping physical artifacts like barns and fences that are visible in the landscape. (p.180)

I take Jackson to mean that, unlike the graffiti investigations, the objects are so noticeable that they force themselves upon the attention of the geographer without any theory to make them stand out or have the geographer go looking for them. The graffiti themselves, while physical configurations, are only meaningful for the mapping process if they are related back to the gang that produced them, and if temporal qualifications, such as order of application or inter-graffiti relationships, are gathered as well. They require a background-theory that allows the association of the tag back to the gang, which is external to the signs.

The alternative approach adopted here is to view culture as the medium or idiom through which meanings are expressed. If one accepts the preceding arguments for a plurality of cultures, then 'culture' is the domain in which these meanings are contested. (p.180)

The situation is even cleared in the case of the mapping of gang violence; though these events and their properties can be localized in space, they no longer persist in space; they are ephemeral and thus require a theory to find them and a GIS to put them into spatial relations. The configuration of the neighborhood, however, which provides the relations of proximity for the analysis of the events, is a separate mapping. Equally separate is the territorial claims of the gangs and their associations. This association may be even more meaningful in situations of collaboration; the Philadelphia setting seems more of a all-against-all configuration, where alliances are not present. Thus the argument is based on a four-way mashup of independent data that is unified across time and space.

1. The incidents of violence, coded as localized gang vs gang expression with a specific severity.
2. The spatial configuration of the neighborhood.
3. The territorial claims of the gangs.
4. The inter-gang alliances (which may be the empty set).

The punchline of Jackson's observation is that, since it is only the models that suggest what data needs to be gathered and thrown onto the maps, the connection to the cultural theories is tantamount for the relevance of cultural geography (p.180).  

This suggests that a model palette is a key resource for the cultural geographer, using distinctions sussed out by cultural theorists in other settings to at least start out from. Here are some of the more interesting models that Jackson depicts in his book.

There is the model of social closure by Frank Parkin, a Weberian redevelopment of Marxist class theory (cf. Frank Parkin, Marxism and class theory: A bourgeois critique, London (Tavistock) 1979) (cf. Jackson's exposition pp.53f; graphic Figure 3.1 on p.55), which is concerned with understanding the way in which dominant groups express their power (hegemony) over subordinate groups. Given the two units, and the uni-directional arrow of dominance, there is either the scenario where the dominant group manages to dominate and project its power over the subordinate group (Parkin terms this exclusionary closure), or the subordinate group manages to exert power over the dominant group (Parkin terms this usurpationary power) [the London Dock Strike of 1889 comes to mind, RCK]. Furthermore, the exclusionary closure can be coupled into a kind of "pecking order" called dual closure, when an intermediate group is identified, which is both dominated by the dominant group but dominates itself over the subordinate group. Jackson gives the example of the Upper and Middle Class dominating the white male working class, which in turn dominates the black & female working class (cf. Figure 3.1 on p.55).

There is also the model of linguistic structuration, as developed by Anthony Giddens (cf. Central Problems in Social Theory, London (Macmillan), 1979; The Constitution of Society, Cambridge (Polity Press) 1985) (p.157). As Peter Jackson explains,

Systems of grammar have no independent existence; they represent the outcome of many individual acts of speech, sedimented through time into rules and conventions. Language is a structure of signification that is reproduced in social practice. Like other practices, however, it does not exist outside social relations of power. Grammatical rules and other linguistic conventions provide a system of sanctions through which certain practices are legitimized and social norms enforced (see Figure 7.1). (p.157)

The cited Figure 7.1 depicts a tabular structure of boxes, where each box is connected to its surrounding neighbors by lines (not arrows). The contents of these boxes runs as follows:

• INTERACTION --- communication --- power --- sanction
• (MODALITY) --- interpretative scheme --- facility --- norm
• STRUCTURE --- signification --- domination --- legitimation 

Unfortunately, Jackson provides no example application, as he had done with the other models or research results that he describes, leaving this schematic somewhat abstract and a mere grab-bag of distinctions that might be employed.

In conclusion, there is an important caveat to raise regarding mapping with respect to the relationship between time and space. In the context of discussing the micro-geography of prostitution (cf. Figure 5.3, p.119), Jackson warns of the danger of the reduction or exclusion of the temporal dimension in the map making static---"freezing" in Jackson's terminology, with scare-quotes in the original---an inherently dynamic process, as "prostitutes themselves adapt their spatial behaviour to prevailing circumstances, constantly moving between and within cities" (p.118). 

For the mapping of graffiti and gang violence, I believe, this would mean that different temporal slices might give different indications of contestedness; furthermore, as gang territories shift or gangs are absorbed, the data has to be adjusted or culled accordingly.

A new research structure

Yesterday evening, my advisor and I tossed the basic setup of the research endeavor into early Mormon history and started over. We kept many of the parts, but he streamlined the process that I will have to go through to simplify the tasks. Of the key topoi that I had presented at the November colloquium, we tossed the question of the Book of Mormon and its connection to Gnostic, Hermetic, Romantic and philosophical concerns, the magical world view and the Second Great Awakening.

Instead, we opted to focus on the financial administration of the young Latter Day Saints congregation; on urbanization as communication in Nauvoo and its temple; and on the general principles of mobility during the time period. It is only when these topics are covered with a "naive" Braudelian tri-partite scheme of temporal structure that we match the interpretative structures to the Mediterranean, eliminating the need to reconstruct models that are unhelpful.

Having thus focused the palette based on contextual use, we can still apply the palette in a fifth chapter to the Mormon War of the 1830s and evaluate its helpfulness and/or omissions and insufficiencies.

With a good 50 pages for each chapter, 20+ pages for Introduction and Conclusion, this puts us at ~300 pages of effort. Furthermore, the approach disentangles the first three chapters on financial administration, urbanization and mobility from the model-theoretic framework, allowing them to stand on their own.

Finally, this approach eliminates a methodological straight-jacket and allows the sources to influence both the findings and the directions of the research, rather than fixing the validity and the schematics of the categories up front.

Scripts versus Ideal Types

On reconsideration, it appears more helpful to treat ideal types as concepts of discovery rather than presentation.
They do capture some of the relations at the type level but the underspecified form makes them more useful for the research phase than for the presentation or description phase.

The precise way in which they are underspecified is their degree of quantification with respect to the relationships.

Upgrading Braudel's model palette

Previously we observed that the conceptual framework for Braudel's Mediterranean needs to be not only extracted, but also updated to reflect past thinking, Braudel's own reconsiderations and more recent thinking on the issues involved by others.

As far as the past thinking is concerned, the odd thing about the Mediterranean is that it's use of historical geography is much less sophisticated than the use is in Lucien Febvre's Introduction géographique (Paris, 1922). This is even more surprising when one considers that Febvre and Braudel did not meet until the 1930s.

Braudel of course had a much larger oeuvre than just the Mediterranean, as often as he returned to the subject matter in revisions and re-scriptions. It would be a caricature of Braudel to not revise his stance from the Mediterranean, where helpful, from the latter works on Capitalism, the identity of France, the History of Civilizations, and various other works. 

Braudel had no direct followers in his approach to historical authorship, but The Corrupting Sea by Horden & Purcell, the richly illustrated The Great Sea: A Human History of the Mediterranean by David Abulafia, and The Middle Sea: A History of the Mediterranean by John Julius Norwich all define and differentiate themselves with respect to Braudel and deserve to influence palette construction.

The "middle duration" (la durée médiane, for want of a better word) was always the step-child of Braudel's model, more illustrated than explicated with the business cycles. But it is in this area of research that population biologist Peter Turchin and his work on clio-dynamics (Secular Cycles; War, Peace and War; Historical Dynamics) has made most relevant contributions, extending the dynamic feedback descriptions from populations to socio-political groups such as artistocracies. I strongly suspect that Braudel would have embraced Turchin's work and found good use for it in yet another rewrite of the Mediterranean. 

The final update that needs to happen is the elimination of any triptych restriction of the temporal stratification. There is no justifiable reason to have just three layers in the historical cake; rather, the number of layers needs to be determined by the temporal aspects of the subject matter at hand and the duration of the various influences. It makes sense to cluster and sort the temporal durations, but there is no advantage to hard-code the cluster size to three. The work on knowledge representation for processes in cell and plant biology, where multiple cycles interlock at different rates and durations to form larger processes, is an example of such interactions.

Building the Model Palette

The construction of the model palette mentioned previously is not restricted to a simple reading of the three volumes of the Mediterranean in various translations, with an eye toward the French single-volume first edition.

The model palette construction process needs to also include a clarification of what the model palette is and what is construction means. This is the place to discuss the difference between scripts (Schank & Abelson) and ideal types (Weber), the problems of mapping plans and goals (Schank & Abelson) onto events (Danto), and the discovery of explanatory rules (Danto again). It is also the place to differentiate oneself from postmodernism and the "large narrative" discussions (Morris), to explicate my personal Kripkensteinian stance, and to accept the relevancy criticism from Danto's demolition of the ideal chronicler (which Braudel seems to not accept for the longue durée, as he pretends to describe the geography independent of the future uses he will put the description to).